Most G protein-coupled receptors, including the mammalian beta(2)-adrenergic receptor, are endocytosed to an intracellular, vesicular compartment upon continued exposure to agonist. The long form of the avian beta(1)-adrenergic receptor, which contains a carboxyl-terminal 59-amino acid extension, does not undergo agonist-promoted endocytosis. We constructed and expressed turkey beta(1)-adrenergic receptor cDNAs with regularly spaced carboxyl-terminal truncations and studied their agonist-promoted endocytosis. Removal of 34-86 amino acids from the carboxyl terminus of the turkey receptor allowed its efficient endocytosis, with optimal endocytosis observed upon re moval of 59 residues. Removal of only 18 residues allowed some endocytosis. A receptor that lacks the entire carboxyl-terminal region (124 residues) was not endocytosed. We also constructed a chimeric hamster beta(2)-adrenergic recep tor with the added 59-residue carboxyl-terminal domain of the turkey receptor. The chimera was not significantly endocytosed. These data indicate that residues 450-465 in the carboxyl-terminal region of the beta(1)-adrenergic receptor can act independently to block agonist-promoted endocytosis and that other carboxyl-terminal structures nearer to the seventh membrane span are required for endocytosis.

In many cells catecholamines induce a translocation of beta-adrenergic receptors from the cell surface to intracellular vesicular sites. We have postulated that the translocation event is the result of ligand-induced endocytosis of the receptor, probably via clathrin-coated pits. Previously, we demonstrated that reduction of cellular ATP content with antimycin A completely blocked endocytosis of epidermal growth factor and translocation of beta-adrenergic receptors in 1321N1 astrocytoma cells. However, the effect of reduction in ATP content on endocytosis remains controversial. In the present report, we demonstrate that reduction of ATP content to a level < 5% of that in control cells is sufficient to prevent endocytosis of [125I]iodotransferrin and translocation of beta-adrenergic receptors. The further demonstration that reactions leading to the return of internalized transferrin or beta-adrenergic receptors to the cell surface are blocked after relatively modest reductions in ATP content provides further evidence of the similarity in the processes subserving diacytosis of beta-adrenergic receptors and transferrin. The differential requirement for ATP of the two arms of diacytosis provides the basis for an explanation of the controversy regarding a requirement for ATP in endocytosis via clathrin-coated pits.

The kinetics of the return of internalized beta-adrenergic receptors to the plasma membrane were measured in human astrocytoma cells. The movement of [125I]iodopindolol-labeled receptors back to the plasma membrane was measured directly and was shown to occur with a t1+2 of 3-4 min. Unlabeled receptors appeared to exhibit the same kinetics of externalization. The process was not inhibited by low concentrations (1-10 microM) of propranolol or even high concentrations of isoproterenol (0.1-1.0 mM). Higher concentrations of propranolol (0.1-1.0 mM) and other lipophilic amines inhibited externalization. The results are consistent with the proposal that catecholamine-induced beta-adrenergic receptor internalization and externalization (diacytosis) occur via the clathrin-coated pit/endosome pathway.

Recombinant turkey erythrocyte beta-adrenergic receptors expressed in murine L cells exhibited characteristic avian subtype selectivity for agonists and antagonists. In 10 of the 11 clones studied, no agonist-induced internalization of receptor was observed, although agonist-induced uncoupling of receptor and adenylyl cyclase occurred rapidly. GTP caused little or no decrease in affinity for beta-adrenergic agonists. Such behavior is commonly observed in avian erythrocytes. In contrast, one clone was susceptible to agonist-induced receptor internalization and down-regulation even though it exhibited characteristic avian beta-adrenergic ligand-binding properties. The affinity of this variant receptor for agonists was also notably reduced by GTP. Electrophoresis of affinity-labeled receptor from this clone indicated an apparent size of about 33 kDa, about 12 kDa less than that of the native or recombinant turkey beta-adrenergic receptor. Genomic DNA from this cell line that encodes the receptor was cloned and partially sequenced. The coding region of the original receptor cDNA was interrupted after codon 412 (out of 483) and was followed by 36 base pairs of novel sequence prior to the first in-frame stop codon. These results suggest that the lack of both hormone-induced internalization and GTP-sensitive, high affinity binding of agonists that is characteristic of the beta-adrenergic receptor in avian erythrocytes is due to intrinsic properties of the receptor. The restoration of these phenomena in a C-terminally truncated mutant receptor suggests the importance of the C-terminal domain in determining these processes.

Receptor-mediated internalization of epidermal growth factor (EGF) occurs by a process involving initially clathrin-coated pits on the cell surface and the subsequent formation of ligand-containing endosomes. Using a modified acid wash technique, cell surface-bound EGF was removed. Utilizing sucrose density centrifugation, the residual cell-associated EGF was separated into plasma membrane-associated and intracellular vesicle-associated forms. Using these procedures we have identified a transient form of cell-associated EGF that is still attached to the plasma membrane but not accessible to the extracellular fluid. This form of EGF appears to be the precursor for endosomic EGF. We suggest that this intermediate form represents the receptor-ligand complex shown by electronmicroscopy to be located in narrow-necked plasma membrane invaginations (Willingham, M. C., and Pastan, I. (1980) Cell 21, 67-77).

Exposure of 1321N1 human astrocytoma cells to the protein kinase C (PKC) activator phorbol 12-myristate, 13-acetate (PMA) led to a rapid and concentration-dependent decrease in isoproterenol (ISO)-stimulated adenylate cyclase (AC) activity in cell lysates. This desensitization of beta-adrenergic receptor (BAR) function was mimicked by mezerein, which also activates PKC, but not by 4-O-methyl-PMA, which is a very weak activator of PKC. Pretreatment with PMA led to desensitization of AC activity stimulated by ISO and by prostaglandin E1, in contrast to the beta-receptor-specific desensitization induced by ISO. Stimulation of AC activity by forskolin and by fluoride remained unaltered. The extent of desensitization observed with PMA plus ISO was greater than with either agent alone. Desensitization with PMA did not result in internalization of BAR, as assessed by sucrose density gradient centrifugation assays and by assays of competition by the hydrophilic ligand ISO for radioligand binding to intact cell receptors. PMA pretreatment did not alter the apparent affinity of the agonist ISO for intact cell BAR, nor was the potency of ISO for stimulation of AC activity altered. The protein kinase inhibitor H7 [1-(5-isoquinolinesulfonyl)-2-methylpiperazine] inhibited the desensitization induced by PMA but not that induced by ISO. These results indicate that activation of PKC can lead to desensitization of receptor-stimulated AC activity but that agonist-induced desensitization of BAR-stimulated AC activity occurs by a different mechanism.

Beta-Adrenergic receptors and epidermal growth factor receptors are both expressed on the cell surface of human astrocytoma cells. Incubation with a catecholamine or epidermal growth factor results in rapid internalization of the respective receptor. The internalized receptors co-migrate in light fractions on sucrose gradients. Astrocytoma cells maintain a constant ATP concentration by either glycolytic or mitochondrial ATP production. When cells are incubated in a medium depleted of substrates for glycolysis and gluconeogenesis, addition of inhibitors of mitochondrial ATP synthesis causes a rapid reduction in cellular ATP content. An immediate return to control ATP levels occurs upon addition of an appropriate nutrient, such as glucose. Decreasing the cellular ATP content to less than 10% of control markedly inhibits internalization of beta-adrenergic receptors and epidermal growth factor. The inhibition of endocytosis is reversed as soon as the intracellular ATP content is restored. Previous work by others (Clarke, B.L., and Weigel, P.H. (1985) J. Biol. Chem. 260, 128-133) suggested that ATP is not required for internalization (per se) of asialoglycoprotein in hepatocytes but was required for recycling of the asialoglycoprotein receptor. In contrast, our results indicate that in astrocytoma cells the process of internalization of epidermal growth factor and beta-adrenergic receptors, per se, is highly ATP dependent.

The redistribution of β‐adrenergic receptors (β‐AR) during agonist‐induced desensitization has been compared to the process of receptor‐mediated endocytosis of epidermal growth factor (EGF) in human astrocytoma cells (1321N1). [ 125 I]EGF exhibited saturable binding to high affinity (K D = 1–2 nM) receptor sites on intact 1321Nl cells. [ 125 I]EGF was found to internalize rapidly using an acid wash technique to remove surface bound hormone. Sucrose density gradient fractionation following exposure to EGF revealed a redistribution of EGF binding sites from high density (heavy peak) to low density (light peak) regions of the gradient. The light peak binding probably represents EGF in internalized vesicles formed during endocytosis. Low temperature (4 °C) or the presence of the lectin concanavalin A (con A) inhibited this ligand‐induced movement of EGF receptors. When cells were incubated simultaneously with EGF and the β‐AR agonist isoproterenol, both receptors were found to co‐migrate in the low density regions of sucrose gradients. No evidence of heterologous ligand‐induced receptor endocytosis was found. These results suggest that the EGF receptors and β‐AR are processed in parallel by 1321N1 cells.

Tunicamycin, which inhibits formation of asparagine-linked glycoproteins, caused a concentration-dependent blockade of beta-adrenergic receptor (beta-AR) accumulation in 1321N1 human astrocytoma cells during growth in culture. A concentration of tunicamycin (0.1 microgram/ml) that inhibited receptor accumulation and [3H]mannose or [3H]glucosamine incorporation into glycoproteins by 90% had only a small effect (10%) on [3H]leucine incorporation into protein, and reduced the rate of cell growth. Incubation in drug-free medium subsequent to treatment of 1321N1 cells with tunicamycin for 48 hr resulted in recovery of beta-AR to control levels within an additional 48 hr. Exposure of cultures to isoproterenol (0.1 microM, 12 hr) caused an 80-90% loss of beta-AR in both pre- and postconfluent cultures; beta-AR recovered to control levels upon removal of isoproterenol. Although both tunicamycin and the protein synthesis inhibitor cycloheximide blocked beta-AR accumulation during growth of 1321N1 cells, neither agent inhibited the appearance of beta-AR during recovery from the down-regulated state in preconfluent cultures. However, cycloheximide, but not tunicamycin, blocked recovery of beta-AR after isoproterenol-induced loss of receptors in postconfluent cultures. In a previous report (Mol. Pharmacol. 26:424-429, 1984), we provided direct evidence that recovery of beta-AR from down-regulation in postconfluent cultures requires de novo synthesis of receptor protein. Thus, the results with tunicamycin are consistent with the idea that recovery of beta-AR in postconfluent cultures requires the synthesis of new beta-AR molecules, but as aglycoproteins that exhibit radioligand-binding characteristics similar to those of native glycoprotein beta-AR.

Two-dimensional polyacrylamide gel electrophoresis and the radioligand (-)-[125I]iodopindolol (125I-Pin) have been used to study isoproterenol-dependent protein phosphorylation and beta-adrenergic receptor availability, respectively, in cultured Sertoli cells and freshly isolated seminiferous tubular segments of sexually immature and mature rats. Sertoli cells prepared from sexually immature rats show progressive 125I-Pin binding in primary cultures that correlates with isoproterenol-induced cell shape changes, redistribution of immunoreactive vimentin, and phosphorylation of this intermediate filament protein. The development of 125I-Pin binding to Sertoli cell lysates is blocked by cycloheximide. Seminiferous tubules do not show significant isoproterenol-dependent vimentin phosphorylation nor 125I-Pin binding. However, vimentin phosphorylation can be induced by follicle-stimulating hormone or a cyclic nucleotide analog. This study stresses the need for correlating pharmacological-induced responses observed in Sertoli cell primary cultures with those in the intact seminiferous tubule.

Competition by beta-adrenergic agonists and antagonists for 125I-pindolol binding sites on intact cells (1321N1 human astrocytoma and C62B rat glioma) was measured using short time binding assays as previously described (Toews, M. L., Harden, T. K., and Perkins, J. P. (1983) Proc. Natl. Acad. Sci. U.S.A. 80, 3553-3557). Preincubation of cells with agonists converted about half of the cellular beta-adrenergic receptors from a form exhibiting high affinity for the agonists isoproterenol and epinephrine and the antagonist sotalol to a form exhibiting much lower apparent affinity for these ligands in short time assays. Exposure to agonists did not alter the affinity of receptors for the antagonist metoprolol. This change in the ligand binding properties of the receptor was rapid (t1/2 = 1-2 min following a lag of about 0.5 min), reversible (t1/2 = 6-8 min), and dependent on the agonist concentration present during the preincubation (K0.5 = 15 nM for isoproterenol). Both isoproterenol and sotalol attained equilibrium with the high affinity receptors very rapidly but equilibrated only slowly with those receptors exhibiting low apparent affinity in short time assays. These results are interpreted in terms of a model which postulates that both the low apparent affinity in short time assays and the subsequent slow equilibration of hydrophilic ligands with these receptors result from agonist-induced internalization of a fraction of cell surface beta-adrenergic receptors. The relationship of this change in receptor binding properties to other aspects of agonist-induced desensitization of the beta-adrenergic receptor-coupled adenylate cyclase system is discussed.

Exposure of postconfluent 1321N1 human astrocytoma cells to 1.0 microM isoproterenol for 12-24 hr results in a 90% loss of beta-adrenergic receptors. Upon removal of agonist, recovery of beta-receptors to control levels occurs within 72 hr. The recovery of receptors is completely blocked by cycloheximide [R. C. Doss, J. P. Perkins, and T. K. Harden, J. Biol. Chem. 256:12281-12286 (1981)]. In contrast cycloheximide does not block recovery of beta-receptors after down-regulation in preconfluent cultures. To determine unambiguously if beta-receptor synthesis accounts for the recovery of receptors after down-regulation, post confluent cultures were incubated with isoproterenol and then transferred to agonist-free medium containing either normal or "heavy" (2H, 13C, 15N) amino acids. The rate and extent of beta-receptor recovery were similar in both normal and heavy amino acid-containing medium. When beta-receptors that had recovered in the heavy amino acid-containing medium were labeled with 125I-cyanopindolol, solubilized in Lubrol PX, and subjected to centrifugation on a 5-15% sucrose density gradient, they exhibited an increased mass compared to beta-receptors that recovered in the presence of normal amino acids. These results confirm that the density shift method is a useful approach for the study of beta-receptor synthesis and that new receptor synthesis occurs during recovery of beta-receptors from catecholamine-induced down-regulation in postconfluent cultures.

beta-Adrenergic receptor (beta AR)-specific, agonist-induced desensitization of adenylate cyclase can be shown in most mammalian cells examined to involve at least three reactions. An initial 'uncoupling' reaction leads to a 40-60% loss of catecholamine-stimulated adenylate cyclase activity at a time when no detectable loss of beta AR has occurred. This process precedes by 45-90 sec the appearance of beta AR in cytoplasmic vesicles. Such beta AR exhibit ligand binding properties consistent with their existence on the inside of membrane vesicles; thus, they appear to be formed by a process of agonist-induced beta AR internalization (endocytosis). A third process results in the loss of beta AR, at least in some cases due to receptor degradation. In general, agonist-induced desensitization or down-regulation reactions do not require protein synthesis. Recovery from the desensitized states does not require protein synthesis, whereas recovery from beta AR down-regulation (degraded receptors) requires new receptor synthesis. Agonist-induced beta AR desensitization and down-regulation reactions appear to have much in common with the process of polypeptide hormone-induced receptor down-regulation. The availability of a large number of ligands (agonists, partial agonists and antagonists) for the beta AR should allow the use of this receptor system to gain unique insights into the general processes of ligand-induced, cell surface receptor endocytosis.

In conclusion, our data indicate that exposure of 1321N1 astrocytoma cells to catecholamines initiates a series of reactions that decreases cellular responsiveness to catecholamines and eventually results in a loss of functional beta AR from the cell. Whether the series of reactions depicted in Fig. 1 indeed represents a sequential process is not yet known. The molecular nature of the beta AR modifications responsible for uncoupling, internalization, and loss of binding also remains unknown. Nonetheless, these reactions are not restricted to the beta AR/AC system of 1321N1 astrocytoma or C62B glioma cells, since a number of investigators working with a variety of homogeneous cell systems have reported the occurrence of similar phenomena (2,7,8,13-16,19,21,28). Thus, the basic framework of catecholamine-induced modification of the properties of beta AR in astrocytoma cells is likely representative of the general phenomenon of agonist-specific desensitization of the beta AR-linked AC of mammalian cells.